On Tue, 17 Oct 2000, Chris Waters wrote:
> On Tue, Oct 17, 2000 at 12:31:18PM +0200, Roland Rosenfeld wrote: > > On Mon, 16 Oct 2000, Chris Waters wrote: > > > > set $(runlevel) # $2 is now current runlevel > > > name=service > > > rcfile=/etc/rc$2.d/S??$name > > > > test -f $rcfile && $rcfile restart > > > > Simple, cleaner, more elegant, more obvious, less confusing. > > > And completely incompatible with file-rc > > Ah, hadn't noticed that we have such a beast. I'll leave aside my > personal feelings about such an abomination and agree that it does > break my proposed example. Fortunately, it was just an example. > > > we haven't a script to find out whether a daemon is running yet, but > > we should introduce one and fixate this in the policy). > > Yes, this would seem to be the only sane approach. (Other than > discarding file-rc and shooting Roland to put him out of his > misery.:-) This is fodder for another thread, anyway. Historical reasons aside, is there a Good Enough valid technical reason to discard file-rc? The only difficulty I've had with using file-rc is with third-party install scripts that assume the system has /etc/rc?.d/<foo> symlinks. Plus I seem to recall Red Hat going off and doing their own thing for a while with /etc/<foo>/rc?.d/ (I just don't recall the exact name they use) Hell, I've been using file-rc ever since it was introduced. Searching the archives, it appears it was with the 2.0 release, and I vaguely recall using it somewhat beforehand. [stuff about the libc upgrades snipped to fork the thread]
